Bile acid biosynthesis during development: hydroxylation of C27-sterols in human fetal liver

Transcription

1 Bile acid biosynthesis during development: hydroxylation of C27sterols in human fetal liver Jan Gustafsson Departments of Pediatrics and Pharmaceutical Biochemistry, Uppsala University, Uppsala, Sweden Abstract Several hydroxylase activities in bile acid biosynthesis were assayed in subcellular fractions of human fetal liver. The livers were obtained at legal abortions performed between gestational weeks 14 and 24. Microsomal l2ahydroxylase and mitochondrial 12a and 26hydroxylase activities were detected from week 14. The microsomal fraction also had capacity for 25hydroxylation, whereas 7a and 26hydroxylase activities were hardly detectable. The variation of the hydroxylase activities between different experiments can be explained by inactivation during the abortion or workup procedure. The results are discussed with respect to earlier studies of bile acid biosynthesis during development and adult life. Gustafsson, J. Bile acid biosynthesis during development: hydroxylation of Cz7sterols in human fetal liver. J Lipid Res : Supplementary key words hydroxylase activities microsomes mitochondria The development of bile acid biosynthesis during fetal life is important for the process of lipid absorption in the newborn (1). Work by Watkins et al. (2) indicates a low capacity for bile acid biosynthesis in newborns, especially in prematures. This may be one explanation for the functional immaturity of neonatal lipid absorption (3). The presence of bile acids in the fetal gall bladder from 14 weeks of gestation (3) indicates the occurrence of fetal bile acid biosynthesis. However, the presence of secondary bile acids shows that at least part of the bile acids is derived from the mother (3). There is little information available concerning individual reactions in fetal bile acid biosynthesis (3). Recently, the mitochondrial fraction of human fetal liver was shown to hydroxylate 5/3cholestane3a,7adiol in the sterol nucleus as well as in the side chain (4). In addition, fetal liver microsomes hydroxylated the same compound in the 12aposition. The present work reports studies of hydroxylations of bile acid precursors in the mitochondrial and microsomal fractions of human fetal liver. Labeled compounds MATERIALS AND METHODS [4 C]Cholesterol (sp act 61 Ci/mol) was obtained from the Radiochemical Centre (Amersham, England). Before use, the material was purified by chromatography on aluminium oxide, grade I11 (Woelm, Eschwege, West Germany) (5). 5[7@SH]Cholestene3/3,7adiol (sp act 5 Cilmol), 7a[6PSH]hydroxy4cholesten3one (sp act 7 Ci/mol), 5~[7/33H]cholestane3a,7adiol (sp act 50 Ci/mol) and 5P[7/3SH]cholestane3a,7a,12atriol (sp act 50 Ci/mol) were prepared as described previously (6). Unlabeled compounds 5Cholestene3P,26diol, 5cholestene3/3,7a,26triol, 7a,l2adihydroxy4cholesten3one, 5/3cholestaneJa, 7a,l2atriol, 5/3cholestane3a,7aY,26triol, and 5Pcholestane3a,7a,l2a,26tetrol were prepared as described previously (6). Enzymes and cofactors NADPH and isocitric acid were obtained from Sigma Chemical Go. (St. Louis, MO). Experimental Human fetuses were obtained at legal abortions performed in weeks 1424 for sociomedical reasons. Fetal age was determined from data concerning the pregnancies and from height measurements. Consent was given by the ethical committee of the University of Uppsala. The abortions were performed by use of prostaglandins or via hysterotomy. The fetuses were taken to the laboratory and liver tissue was taken out and chilled in icecold buffer solution. Preparation of liver homogenates was started within 4560 min after abortion. Liver homogenates (low, w/v) were prepared in 0.25 M sucrose with a PotterElvehjem homogenizer equipped with a loosely fitting pestle (6). The microsomal fraction was obtained from the homogenate by centrifugations at 800 g, 20,000 g, and 100,000 g (6). The microsomal pellet was finally suspended and homogenized in 0.05 M Trisacetate, ph 7.4, in a volume corresponding to 1/4 to 1/2 of the initial volume. The mitochondrial fraction was obtained by resuspension in 0.25 M sucrose of the pellet obtained after centrifugation at 20,000 g and recentrifugation twice at 6,400 g for 20 min (7). The final precipitate was sus Journal of Lipid Research Volume 27,

2 pended in 0.1 M TrisC1 buffer, ph 7.4, in a volume corresponding to 1/6 to 1/4 of the initial volume of the homogenate. The protein content of the microsomal fraction was between 0.5 and 3.0 mg/ml and that of the mitochondrial fraction between 0.4 and 2.5 mg/ml when determined according to Lowry et al. (8). The lack of detectable 25hydroxylase activity towards 50cholestane 3a,7a,12atriol in the mitochondrial fraction as well as lack of 26hydroxylase activity towards the same substrate in the microsomal fraction (c.f. Results) speak against occurrence of a significant contamination with microsomal protein in the mitochondrial fraction and vice versa. Both the microsomal and mitochondrial fractions were contaminated to some extent with peroxisomal protein as judged from measurements of the activity of urate oxidase, a peroxisomal marker enzyme (9). In an experiment with a liver from a 15weekold fetus, about 75% of the total urate oxidase activity sedimented with the 800g pellet, about 15% with the mitochondrial fraction, and about 10% with the microsomal fraction. In one experiment the effect on microsomal and mitochondrial Cz7sterol hydroxylation of preincubation of whole liver at 37OC was studied. Thus, a liver from an 18weekold fetus was divided into three parts. One of these was immediately used for preparation of subcellular fractions, whereas the other two were incubated at 37OC for 45 and 90 min, respectively, before preparations of subcellular fractions were started. In another experiment the hepatic concentrations of cholic and chenodeoxycholic acid were determined by isotope dilutionmass spectrometry after addition of deuteriumlabeled internal standards (lo), alkaline hydrolysis, acidification, and ether extraction of liver tissue from two 18weekold fetuses. The concentrations of cholic acid were 3.2 and 4.2 nmol per g of liver, respectively, and those of chenodeoxycholic acid were 1.7 and 4.1 nmol per g of liver, respectively. The hepatic concentration of deoxycholic acid was less than 1 nmol per g of liver. Incubation procedures and analysis of incubation mixtures In incubations with the microsomal fraction, 0.25 pmol of substrate dissolved in 50 pl of acetone was incubated with 2 ml of microsomal fraction in a total volume of 3 ml of homogenizing medium. The incubation mixtures were supplemented with 3 pmol of NADPH. Microsomal incubations with cholesterol were performed as described earlier (11). In incubations with the mitochondrial fraction, 0.25 pmol of substrate dissolved in 50 pl of acetone was incubated with 2 ml of mitochondrial fraction in a total volume of 3 ml of the homogenizing medium. The incubation mixtures were supplemented with 3 pmol of NADPH and 30 pmol of MgC12 (5). In some experiments NADPH was replaced by 4.6 pmol of isocitric acid (5). All incubations were performed for 20 min at 37OC and were terminated by the addition of 5 ml of 96% (v/v) ethanol. After acidification and ether extraction, the incubations were subjected to thinlayer chromatography using appropriate external standards (6). Radioactivity in the different chromatographic zones was assayed with a thinlayer scanner (Berthold, Karlsruhe, Germany) (12). The chromatographic zones corresponding to the products were extracted with methanol, converted into trimethyl silyl ethers (12), and analyzed by radiogasliquid chromatography using a BarberColman 5000 instrument equipped with a 3% QF1 column (12). Knowing the specific radioactivity of the compounds used, the conversion into different products was calculated from the amount of radioactivity found in the thinlayer chromatographic zone and the amount of radioactivity found with the appropriate retention time in the radiogas chromatogram. As judged from gasliquid chromatography of extracts of microsomal and mitochondrial fractions, there was no endogenous dilution with any of the' incubated Cp7sterols except for cholesterol. In case of microsomal and mitochondrial hydroxylations of cholesterol, 50cholestane3a,7adiol and 50cholestane 3a,7a,l2atriol as well as mitochondrial 25hydroxylation of 7ahydroxy4cholest3one, the identities of the products were confirmed by gasliquid chromatographymass spectrometry (6) or by crystallization to constant specific radioactivity after dilution with unlabeled authentic material (cf. ref. 4). RESULTS Incubations with the microsomal fraction Out of 15 tested livers, microsomal hydroxylase activities were detected in 11. In the active preparations the relative conversions of the compounds varied between 0.1 and 4%. The variation between duplicate incubations from the same liver was less than 15%. Table 1 summarizes the hydroxylase activities and the number of metabolic products in incubations of different substrates with the microsomal fraction of livers from fetuses of different gestational ages. Due to the small amounts of liver tissue available, only a limited number of experiments could be performed with each preparation. Microsomal hydroxylase activity was detected in gestational week 14. Regardless of fetal age, very low or no conversion was noted in the case of microsomal 7ahydroxylation of cholesterol. With microsomes from one out of four livers, there was some 12a and 26hydroxylation of 5cholestene30,7adiol. In experiments with 7ahydroxy4cholesten3one, some degree of 12ahydroxylation was obtained. With one liver from week 24 there also was some 26hydroxylation. 56 Cholestane3a,7adiol was also hydroxylated predomi 802 Journal of Lipid Research Volume 27, 1986

3 ~~ TABLE 1. Hydroxylation of CZIsterols by the microsomal fraction of fetal liver Gestational Week aHydroxylation Cholesterol 1 PaHydroxylation 5Cholestene38,7adiol 7aHydroxy4cholesten3one 58Cholestane3a,7adiol 25Hydroxylation 58C holestane3a, 7adiol 58Cholestane3a,7a, 12atriol 26Hydroxylation 5Cholestene38,7adiol 7aHydroxy4cholesten3one 58Cholestane3a, 7adiol 58Cholestane3a,7au, 1 Patriol pmol x mg prot" x min' < Not determined. nantly in the l2aposition. Almost no 26hydroxylation was obtained with this substrate. The product in incubations with 5/3cholestane3a,7a,l2atriol was identified as 5/3cholestane3a,7a,l2a,25tetrol. There was almost no 26hydroxylation of this substrate. Incubations with the mitochondrial fraction Out of 15 tested livers, it was possible to detect one or several mitochondrial hydroxylase activities in 11. In the active preparations the relative conversions of the compounds varied between 0.1 and 6%. The variation when duplicate incubations from the same liver were performed was less than 15%. Table 2 summarizes the hydroxylase activities and the number of metabolic products in incubations of different TABLE 2. substrates with the mitochondrial fraction. Very low 26hydroxylase activity towards cholesterol was found. Only 2 out of 11 preparations were active towards this substrate. In one experiment with a liver from a fetus in week 24, cholesterol was hydroxylated in the (225 position to a small extent. Mitochondrial hydroxylation of 5cholestene3/3,7adiol was tested with two preparations from livers of fetuses in week 24. Both 12a and 26hydroxylations were obtained. Mitochondrial hydroxylation of 7ahydroxy4cholesten3one was studied with two preparations. One of these, a preparation from the liver of a fetus in week 16, yielded only the 25hydroxylated product, whereas the other preparation from a fetus in week 24 had 12a as well as 26hydroxylase activity. Mitochondrial 12a and 26hydroxylation of 5/3cholestane Hydroxylation of Cz7sterols by the mitochondrial fraction of fetal liver Gestational Week pmol x 1 PaHydroxylation 5Cholestene38,7adiol 7aHydroxy4cholesten3one 5@Cholestane3a, 7adiol 25Hydroxylation Cholesterol 7aHydroxy4cholesten3one 26Hydroxylation Cholesterol 5Cholestene3@,7adiol 7aHydroxy4cholesten3one 5flCholestane3aI 7adiol 58Cholestane3a, 7a, 1 Patriol < "Not determined. Gusfufson Hydroxylation of Cz7eterols in fetal liver 803

4 ~~~ ~ 3a,7adiol occurred in four out of eight livers. The activities varied markedly between different liver preparations. Irrespective of fetal age, incubations of 50cholestane 3a,7a,12atriol gave only 26hydroxylated product. This was obtained in ten out of ten livers. In two experiments, incubation of 5~cholestane3a,7ar,12atriol with the mitochondrial fraction from fetuses in weeks 16 and 24, gave in addition to 5~cholestane3a,7a,12~.~,26tetrol also small amounts of a product with thinlayer chromatographic properties of 3a,7a,12atrihydroxy5/3cholestanoic acid. Effect on microsomal and mitochondrial hydroxylations of preincubation of whole liver at 37OC When whole liver from an 18weekold fetus was incubated at 37OC for 45 or 90 min, microsomal 25hydroxylation of 5Pcholestane3a,7a,l2atriol was inhibited by about 40% and mitochondrial 26hydroxylation of 5Pcholestane3a,7a,l2atriol by about 70 and 80%, respectively (Table 3). Effect on microsomal and mitochondrial hydroxylations of bile acids There was no inhibition of microsomal 25hydroxylation or mitochondrial 26hydroxylation of 5Pcholestane 301,701,12atriol when preparations from the liver of an 18weekold fetus were incubated with the addition of taurocholic acid, 2.1 or 4.2 pm, and taurochenodeoxycholic acid, 2.1 or 4.2 pm, respectively (Table 4). DISCUSSION Cytochrome P450dependent, microsomal hydroxylations in human fetal liver have been shown with laurate (13), drugs (14), and several steroids (15). In addition to recent work on 12ahydroxylation of 56cholestane 3a,7adiol in fetal liver (4), the present work shows that the microsomal and mitochondrial fractions of fetal liver have capacity for several hydroxylations of C2,sterols that TABLE 3. Effect of preincubation of fetal liver at 37OC on microsomal and mitochondrial hydroxylations Percentage Relative Conversion at Time of Preincubation 0 min 45 mln 90 min Microsomal 25hydroxylation 5flcho1estane3a,7ay, 1 Patriol Mitochondrial 26hydroxylation 5flcholestane3a,7a, 12atriol The experiment was performed with liver tissue from an 18weekold fetus. The incubations were performed as described in Methods. The conversion rate obtained with material that was not preincubated was set to 100%. are intermediates in bile acid synthesis in adult liver. There was no clear increase in hydroxylase activities with gestational age. In evaluating the efficiency of conversion, one has to consider the possibility of partial inactivation of the enzymes during abortion or during preparation of the subcellular fractions. Evidence for the possible occurrence of such inactivation was obtained in an experiment in which whole fetal liver was incubated at 37OC before preparation of subcellular fractions. However, in the overall investigation, the microsomal and/or the mitochondrial fractions of 13 out of 15 livers were active towards one or more of the substrates tested. Microsomal hydroxylations Regardless of fetal age, almost no microsomal 7ahydroxylase activity towards cholesterol was detectable. Since microsomal 7ahydroxylation of cholesterol is the ratelimiting step in bile acid synthesis in adult liver (l), the low degree of 7ahydroxylation of cholesterol may support findings by Watkins et al. (2) of low bile acid pools in newborns, especially prematures, as compared to adults. 7aHydroxylation of taurodeoxycholic acid was recently shown to occur in fetal human liver microsomes (16). At least in rat liver, this reaction is catalyzed by a species of cytochrome P450 different from that catalyzing cholesterol 7ahydroxylation (17). In similarity with in vitro preparations of adult human liver, major microsomal hydroxylations of fetal liver are 12a and 25hydroxylation (12). The effect of addition of conjugated cholic and chenodeoxycholic acid, in concentrations occurring in liver, on mitochondrial and microsomal hydroxylation of 50 cholestane3a,7a,l2atriol was tested. Since taurine conjugates predominate in the fetus (3), taurocholic and taurochenodeoxycholic acid were used. Neither of these bile acids had any inhibitory effect on the hydroxylations tested. This is not consistent with an inhibitory effect of bile acids on hydroxylations of CP7sterols in fetal liver. However, in view of the characteristics of the adult 7ahydroxylase activity in vivo (l), it cannot be excluded that the low 7ahydroxylase activity in the present study is due in part to an inhibition by intrahepatic bile acids. Due to the sedimentation properties of fetal liver subcellular fractions, hydroxylase activities can be found already in the pellet obtained after centrifugation at 800 g (18). This was confirmed by assay of CZ7sterol hydroxylase activities in the 800 g pellet in some experiments. In relation to the amount of subcellular protein incubated, the microsomal hydroxylases were less active than the mitochondrial hydroxylases. This differs from what is found in adult liver preparations (12). Mitochondrial hydroxylations The low extent Of 26hydroxy1ation Of cholesterol under the conditions used is surprising since 26hvdroxycholes 804 Journal of Lipid Research Volume 27, 1986

5 TABLE 4. Effect of addition of taurocholic (TCA) and taurochenodeoxycholic (TCDA) acid on microsomal and mitochondrial hydroxylations of 5j3cholestane3a,7a, 12atriol TCA + No Addition TCDA 2.1 p~ TCA 4.2 PM + TCDA Microsomal 25hydroxylation Mitochondrial 26hydroxylation percentage relative conversion The experiment was performed with liver tissue from an 18weekold fetus. The conversion rate in the incubations without any addition of bile acids was set to 100%. terol has been found in meconium (19) and 30hydroxy5 cholenoic acid has been found in both meconium and amniotic fluid (20, 21). 30Hydroxy5cholenoic acid has its origin in 26hydroxycholesterol and is supposed to represent a primitive pathway for synthesis of monohydroxy bile acids in the fetus (21). On the other hand, Makino et al. (22) have shown that cholesterol was rather poorly converted into 3/3hydroxy5cholenoic acid in a child with biliary atresia. The authors suggested that another precursor than cholesterol itself might be involved in this particular transformation (22). In the present investigation, the low 26hydroxylase activity towards cholesterol may well be due to partial inactivation of the 26hydroxylase system. The lability of the system has been pointed out earlier both in work with adult human liver and rat liver (12, 23). In one experiment some degree of mitochondrial 25hydroxylation of cholesterol occurred and in another 25hydroxylation of 7ahydroxy4cholesten3one. 25 Hydroxylation of cholesterol has earlier been shown with rat (5) but not with human liver mitochondria (12). The importance of 25hydroxylation of cholesterol in vivo is questionable since, at least in adults, 25hydroxycholesterol is a poor precursor of bile acids (24). Liver mitochondrial 25hydroxylation can be important in vitamin D activation (25), but it remains to be established whether fetal liver mitochondria has capacity for vitamin D 25 hydroxylation. Pathways in fetal bile acid synthesis The results of the present work are consistent with fetal bile acid synthesis occurring along pathways similar to those in adult human liver (1). In spite of the low degree of 7ahydroxylation of cholesterol under the in vitro conditions used (ll), it is still likely that this reaction initiates fetal bile acid synthesis in vivo. No other microsomal hydroxylation of cholesterol did in fact occur, making any other microsomal hydroxylation of cholesterol in vivo unlikely. The present results do not rule out the possibility that mitochondrial 26hydroxylation of cholesterol could initiate part of the fetal bile acid synthesis. Despite the low conversions obtained in the present investigation, the oc currence of fetal 26hydroxycholestero1(19) and associated compounds (20, 21) makes it probable that some mitochondrial 26hydroxylation of cholesterol occurs in vivo. To some extent, fetal 26hydroxycholesterol could be derived from the mother by placental transfer, since the compound has been shown to occur in human serum (26). As in adult liver, l2ahydroxylation probably may occur with different substrates. In this work, 5cholestene 3/3,7adiol, 7ahydroxy4cholesten3one, were shown to be l2ahydroxylated. The relative roles of the microsomal and mitochondrial l2ahydroxylase activities cannot be assessed from the present work. The mitochondrial 12ahydroxylase activity is probably lost during the further ontogeny (4), since it has not been found in vitro in preparations from livers of healthy adults (12, 27). As mentioned before, there was no evidence of any significant contamination with mitochondrial protein in the microsomal fraction or vice versa. The small amounts of fetal liver tissue available have prevented the preparation of a peroxisomeenriched fraction. Recent work by Thompson and Krisans (28) has provided evidence for the presence of peroxisomal side chain hydroxylase activity towards 5@cholestane3a,7a,l2atriol in rat liver. As judged from determinations of urate oxidase activity (9), the microsomal and mitochondrial fractions in the present investigation were contaminated with peroxisomal protein to about the same extent. This contamination was evidently not associated with any detectable peroxisomal side chain hydroxylase activity, neither in the microsomal nor in the mitochondrial fraction. If this had been the case, one would have expected the same stereospecific hydroxylations of 50cholestane 3a,7a,l2atriol to occur to some extent in both subcellular fractions. Mitochondrial 26hydroxylation in fetal bile acid biosynthesis may probably occur with several substrates. In cholic acid synthesis, 5Bcholestane3a,7a,12atriol may be a substrate in vivo, whereas in chenodeoxycholic acid synthesis 26hydroxylation of several dioxygenated sterols is possible. 7a,26Dihydroxy4cholesten3one was shown recently to be efficiently converted into chenodeoxycholic acid in vivo in the adult human (29). Probably the fetal liver also has capacity for further oxidation of the sterol Gurtofsson Hydroxylation of Cz7sterols in fetal liver 805

6 side chain, since 3a,7a,12atrihydroxy5/3cholestanoic acid was tentatively identified in two mitochondrial incubations with 50cholestaneSa!,7a,12a!triol. In cholic acid synthesis in the adult, a pathway involving 25hydroxylation may exist (30). Such a pathway involves 24Phydroxylation of 5~cholestane3a,7a!,12a,25tetrol. Although no microsomal 24Phydroxylation occurred under the present conditions, further studies are needed to evaluate the possibility of a 25hydroxylation pathway in fetal liver. I The skillful technical assistance of Mrs. Kerstin Ronnquist and Miss Asa Bystrom is gratefully acknowledged. The present work was supported by the Swedish Medical Research Council (project 03X218), the Expressen Prenatal Research Foundation, and the Swedish Society of Medical Sciences. Manuscrip rccciued 3 January REFERENCES 1. Danielsson, H., and J. Gustafsson Biochemistry of bile acids in health and disease. In Pathobiology Annual. Vol. XI. H. L. Joachim, editor. Raven Press, New York Watkins, J. B., D. Ingall, P. Szczepanik, P. D. Klein, and R. Lester Bile salt metabolism in the newborn. Measurement of pool size and synthesis by stable isotope technique. New Engl. J. Med. 288: Watkins, J. B., and J. A. Perman Bile acid metabolism in infants and children. Clin. Gastmentml. 6: Gustafsson, J Bile acid synthesis during development. Mitochondrial 12ahydroxylation in human fetal liver. J Clin. Invest. 75: Bjorkhem, I., and J. Gustafsson Mitochondrial o hydroxylation of cholesterol side chain. J. Biol. Chem. 249: Bjorkhem, I., and J. Gustafsson ohydroxylation of steroid sidechain in biosynthesis of bile acids. Eur. J. Biochcm. 36: Sottocasa, G. L., B. Kuylenstiema, L. Ernster, and A. Bergstrand Separation and some enzymatic properties of the inner and outer membranes of rat liver mitochondria. Methodr Entymol. 10: Lowry, 0. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193: Schneider, W. C., and G. H. Hogeboom Intracellular distribution of enzymes. IX. Certain purinemetabolizing enzymes. J. Biol. Chem. 195: Bjorkhem, I., and 0. Falk Assay of the major bile acids in serum by isotope dilutionmass spectrometry. Scand. J. Clin. Lab. Invest. 43: Gustafsson, J Effect of biliary obstruction on 26hydroxylation of Cp7steroids in bile acid synthesis. J. Lipid Res. 19: Bjorkhem, I., J. Gustafsson, G. Johansson, and B. Persson Biosynthesis of bile acids in man. Hydroxylation of the C,,steroid side chain. J. Clin. Invest. 55: Yaffe, S. J., A. Rane, E Sjoqvist, L0. BorCus, and S. Orrenius The presence of a monooxygenase system in human fetal liver microsomes. L$e Sci. 9: Kitada, M., and T. Kamataki Partial purification and properties of cytochrome P450 from homogenates of human fetal livers. Biochem. Phamol. 28: IngelmanSundberg, M., A. Rane, and JA. Gustafsson Properties of steroid hydroxylases in human fetal liver. Biochemistry. 14: Gustafsson, J., S. Andersson, and J. Sjovall Bile acid metabolism during development. Metabolism of taurodeoxycholic acid in human fetal liver. Biol. Neonak. 47: Hansson, R., and K. Wikvall Hydroxylations in biosynthesis and metabolism of bile acids. Catalytic properties of different forms of cytochrome P450. J. Biol. Chem. 255: Ackermann, E., A. Rane, and J. L. E. Ericsson The liver microsomal system in the human fetus: distribution in different centrifugal fractions. Clin. Phannacol. Therapcut. 13: Lavy, U., S. Burstein, M. Gut, and N. B. Javitt Bile acid synthesis in man. 11. Determination of 7ahydroxycholesterol, (22R)22hydroxycholesterol, and 26hydroxy cholesterol in human meconium. J. Lipid Res. 18: Back, P., and K. Ross Identification of 3Phydroxy 5cholenoic acid in human meconium. HoppeZeylr s Z. Physiol. Chem. 354: Delirze, G., G. Karlaganis, W. Giger, M. Reinhard, D. Sidiropoulos, and G. Paumgartner Identification of 3/3hydroxy5cholenoic acid in human amniotic fluid. In Bile Acid Metabolism in Health and Disease. G. Paumgartner and E. Stiehl, editors. MTP Press Limited, Lancaster Makino, I., J. Sjovall, A. Norman, and B. Strandvik Excretion of 3Phydroxy5cholenoic and 3ahydroxy5acholanoic acids in urine of infants with biliary atresia. FEBS Lett. 15: Taniguchi, S., N. Hoshita, and K. Okuda Enzymatic characteristics of COsensitive 26hydroxylase system for 5/3cholestane3a,7a,l2atriol in rat liver mitochondria and its intramitochondrial localization. Eur. J. Biochem. 40: Swell, L., C. Schwartz, J. Gustafsson, H. Danielsson, and Z. R. Vlahcevic A quantitative evaluation of the conversion of 25hydroxycholesterol to bile acids in man. Biochim. Biophys. Acta. 663: Bjorkhem, I., and I. Holmberg Assay and properties of a mitochondrial 25hydroxylase active on vitamin Ds. J. Biol. Chem. 253: Javitt, N. B., E. Kok, S. Burstein, B. Cohen, and J. Kutscher Hydroxycholesterol. Identification and quantitation in human serum. J. Biol. Chem. 256: Oftebro, H., I. Bjorkhem. F. C. Stprmer, and J. I. Pedersen Cerebrotendinous xanthomatosis: defective liver mitochondrial hydroxylation of chenodeoxycholic acid precursors. J. Lipid Res. 22: Thompson, S. L., and S. K. Krisans Evidence for peroxisomal hydroxylase activity in rat liver. Biochm. Biophys. Res. Commun. 130: Swell, L., J. Gustafsson, C. C. Schwartz, L. G. Halloran, H. Danielsson and Z. R. Vlahcevic An in vivo evaluation of the quantitative significance of several potential pathways to cholic and chenodeoxycholic acids from cholesterol in man. J. Lipid Res. 21: Shefer, S., F. W. Cheng, B. Dayal, S. Hauser, G. S. Tint, G. Salen, and E. H. Mosbach A 25hydroxylation pathway of cholic acid biosynthesis in man and rat. J. Clin. Invest. 57: Journal of Lipid Research Volume 27, 1986